This article was downloaded by: [123.136.112.

208]
On: 29 June 2015, At: 09:51
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Educational Action Research

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/reac20

Kurt Lewin and the Origins of Action Research

a
Clem Adelman
a
University of Reading , United Kingdom
Published online: 11 Aug 2006.

To cite this article: Clem Adelman (1993) Kurt Lewin and the Origins of Action Research, Educational Action Research, 1:1,
7-24, DOI: 10.1080/0965079930010102

To link to this article: http://dx.doi.org/10.1080/0965079930010102

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content) contained in the
publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations
or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any
opinions and views expressed in this publication are the opinions and views of the authors, and are not the
views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be
independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses,
actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever
caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Educational Action Research, Volume 1, No. 1, 1993

Kurt Lewin and the

Origins of Action Research

CLEM ADELMAN
University of Reading, United Kingdom
Downloaded by [123.136.112.208] at 09:51 29 June 2015

Kurt Lewin is often referred to as the originator of action research although

he is probably better known as the social psychologist who devised the 'field
theory' of concepts otherwise known as topographical psychology. He and
his early associates in the USA promulgated most of the conceptual
structure of interactive theories of organisational behaviour and enlivened
social psychology. Lewin's many outstanding PhD students were prepared
on the grounds of their knowledge of psychology and social psychology, and
action research was one way to apply some of the psychological ideas to a
practical endeavour. Of the many former students and associates, those
who made a contribution to the testing and development of action research
include Argyris, Bennis, Benne, Cory, Jacques, Lippitt, Marrow and White.
In the late 1930s Kurt Lewin and his students conducted
quasi-experimental tests in factory and neighbourhood settings to
demonstrate, respectively, the greater gains in productivity and in law and
order through democratic participation rather than autocratic coercion.
Lewin not only showed that there was an effective alternative to Taylor's
'scientific management' but through his action research provided the details
of how to develop social relationships of groups and between groups to
sustain communication and co-operation. To achieve such conditions and
relationships required forms of leadership quite different from those
purveyed by the literal followers of Taylor and the misinterpretation of Tyler
which led to a link with Watsonian behaviourism and thus 'behavioural
objectives'. One of the best known summaries of the forms of leadership is
by two of Lewin's former students Cartwright & Zander (1953). Action
research was the means of systematic enquiry for all participants in the
quest for greater effectiveness through democratic participation.
Lewin was particularly concerned to raise the self-esteem of minority
groups, to help them seek "independence, equality, and co-operation-
through action research and other means (Lewin, 1946). He wanted

7
 CLEMADELMAN
minority groups to overcome the forces of 'exploitation' and colonialisation
that had been prominent in their modern histories. He espoused the use of
social science as a means to help solve social conflicts and considered that
the clarification of hypothetical, 'if so', questions was fundamental to all
social science research which for Lewin included action research (Lewin,
1946).
Action research gives credence to the development of powers of
reflective thought, discussion, decision and action by ordinary people
participating in collective research on "private troubles" (Wright Mills, 1959)
that they have in common. That was how Kurt Lewin (1890-1947), whose
first ideas on what he called 'action research' were set out in about 1934
(Marrow, 1969), came to describe its characteristics after a series of
practical experiences in the early 1940s. "No action without research; no
research without action", Lewin concluded.
Lewin had fled Berlin in 1933 taking up a temporary position in the
home economics department at Cornell University and then moving to
psychology at the University of Iowa. His initial attempt to establish a
programme of action research was to propose a Psychological Institute of
Downloaded by [123.136.112.208] at 09:51 29 June 2015

the Hebrew University to seek "the wisest solutions and the best practical
administrative alternatives" (Marrow, p. 81), in order to develop better
communities by helping the new immigrants to Palestine to adjust and
thrive in their new environment. His efforts in this regard did not come to
fruition, notwithstanding that his sponsorship included Eleanor Roosevelt,
J o h n Dewey, Edward Thomdike, Frank Boas, and other outstanding
American academics and philanthropists.(l]
The immediate concern of Jewish philanthropy was to help Jews
escape from Nazi-occupied Europe. Ideas like the Psychological Institute
were given little priority at that time; sufficient funding was not
forthcoming. However, opportunities to explore the possibilities of
community action research did arise subsequently in the USA
Whilst at the University of Iowa, Lewin was invited to work as a
consultant to the Harwood factory in Virginia; Marrow was the managing
director. The factory was newly opened and it was found to be difficult to
recruit skilled workers. Three hundred unskilled trainees, mainly local
women, h a d been employed. There was considerable prejudice amongst the
predominantly female managers towards the view that the trainees would
not be able to do the tasks fast enough or to the same standard. After 12
weeks of training the new employees produced only half as much as
apprentices doing similar tasks in northern US factories. In addition,
morale within the factory was low.
Lewin and his principal co-worker, Alec Barvelas, took part of the new
workforce and divided it into two groups. The first received direct training
given didactically with little opportunity to raise questions. The second
group was encouraged to discuss and decide on the division of tasks and
comment on the training that was given. Over several months the
productivity of the second group was consistently higher than that of the
first. The staff of the second group leamt the tasks faster and their morale

8
 THE ORIGINS OF ACTION RESEARCH
remained high, whereas In the first group morale remained low. This initial
field experiment seemed to vindicate Lewin's observations and belief in
democratic rather than autocratic workplaces. The problem of social
relationships and efficiency in Industry h a s been troubling Lewln since the
early 1920s, marked by a critical paper on Taylorism (Lewln, 1920). The
influence of Lewin's work on Industrial relations h a s been enormous
throughout the world as several of those Interviewed by Marrow for the
biographical volume attested. It was part of Lewin's Insight that he could
take contentious social issues and refute the taken-for-granted, often
pessimistic assumptions about 'human nature', and replace these with
what has become a new 'common sense'.
Action research for Lewin was exemplified by the discussion of
problems followed by group decisions on how to proceed. Action research
m u s t include the active participation by those who have to carry out the
work in the exploration of problems that they identify and anticipate. After
investigation of these problems the group makes decisions, monitoring and
keeping note of the consequences. Regular reviews of progress follow. The
group would decide on when a particular plan or strategy had been
Downloaded by [123.136.112.208] at 09:51 29 June 2015

exhausted and fulfilled, come to nothing, and would bring to these

discussions newly perceived problems.
The experiment at the Harwood plant was inspired by earlier work on
the relationships between autocracy and democracy in the workplace
conducted with Lewin's students, Lippitt & White (1939). However, it was
not until j u s t after Lewin's death in 1947 that the opportunity arose at the
Harwood plant for what seems the definitive action research on the efficacy
of democratic group decision-making in industry. I quote from Marrow:
French aided by Lester Coch, the personnel manager, was able to
cany out the experiment as planned. The investigation called for
introducing the required changes in jobs in three different ways,
each involving a different degree of employee collaboration in
working out details of the proposed new job assignments.

The first group did not participate in any way: the workers were
told to the changes in their jobs, and the production department
explained the new piece [wage] rate. The second group was asked
to appoint representatives to meet with management to consider
methods, piece rates and other problems created by the job
changes. The third group consisted of every member of the unit -
not just the representatives. They met with management, took an
active part in detailed discussions about all aspects of the
change, made a number of recommendations and even helped
plan the most efficient methods for doing the new Job.

The differences in outcome of the three procedures were clear-cut

and dramatic. Average production in the non-participation group
dropped 20 per cent immediately and did not regain the

9
 CLEMADELMAN

pre-change level None per cent of the group quit Morale fell
sharply, as evidenced fay marked hostility toward the supervisor,
by slowdowns by complaints to the union and by other instances
of aggressive behaviour.
The group which participated through representatives required
two weeks to recover its pre-change output Their attitude was
co-operative and none of the members of the group were in sharp
contrast to those in the non-participating group. It regained the
pre-change output after only two days and then climbed steadily
until it reached a level about 14 per cent above the earlier
average. No one quit; all members of the group worked well with
their supervisors and there were no signs of aggression.
French concluded that:
The experiment showed that the rate of recovery is directly
proportional to the amount of participation and that the rates of
Downloaded by [123.136.112.208] at 09:51 29 June 2015

turnover and aggression are inversely proportional to the amount

of participation.
Lewin had said that the constancy of the level of production at Harwood or
at any similar plant could be viewed as a quasi-stationery process in which
two types of forces are in gear: those component forces pushing production
in a downward direction and those pushing production up. The difference
in the strength of these forces makes the difference of production level
between the participating and the non-participating groups.
To those expecting accounts of action research to emulate a case
study, this exemplification of Lewin's work does not leave the interpretation
to the reader. Indeed Lewin and his colleagues framed their interpretations
in the form of scientific axions. Although Lewin's understanding of science
was strongly Informed by his professor, Ernst Cassirer, the onus on
empirically testable propositions as the vindication of expenditure on
research whatever the paradigm, was strongly evidence In his and his
colleagues' reports and in their valuing of 'experimental' action research
above the three other approaches they identified (see below). This does not
detract from Lewin's principles and procedures for co-operative action
research as a means of enquiry specially suited to democratic participation.
However, Lewin's ideas on democratic participation In the workplace did not
Include any critique of the wider society, particularly the range of economic
relations between worker and employer, capital and labour. Indeed a fair
observation would be that although Lewin and his co-workers demonstrated
the efficacy of action research for improving productivity, they did not
develop conceptual structures that took explicit account of the power bases
that define social roles and strongly influence the process of any change in
the modes of production. In the context of industrial management the
criticisms of Landsberger (1958) are precise and pertinent:

10
 THE ORIGINS OF ACTION RESEARCH

Equally astonishing in the freedom from attack enjoyed so far by

the followers of Lewin and the group dynamics approach. Coch
and French's action research: 'Overcoming Resistance to Change'
is far more blatant in accepting management's goal of efficiency,
and the desirability of manipulating workers than any study ever
undertaken by the follower of the late Elton Mayo.
In the context of progressive education, as espoused by John Dewey and
George Counts, Lewin had developed the methods and principles to enable
the school to act as the agency of democratic change within its community.
Lewin and Dewey met and corresponded briefly on a few occasions. I have
yet to locate any record of Lewin knowing of the contemporaneous work of
Ralph Tyler at Ohio University during the eight-year study (1932-40).
However, the resemblance between action research and the 'service' studies
by groups of teachers into their own practices that Tyler developed is
uncannily close (Madaus & Stufflebeam, 1989). Tyler had studied Dewey
and they were subsequently colleagues at Chicago University. Myles Horton
(Kohl & Kohl 1990), a major figure in education for adult empowerment,
Downloaded by [123.136.112.208] at 09:51 29 June 2015

corresponded with Lewin and Dewey but did not, at that time, know of
Tyler. Horton contended that action research was too esoteric for working
people (Kohl & Kohl, 1990).
The context for understanding Horton's reluctance is to be found in
Lewin's frank admittance that 'community councils' did not discriminate
between the democratic aims of social science as advocated by Lewin from
the social science of the 'technocracy' of which they had prior experience;
the latter as an autocratic arm of central government informed by university
researchers.
The community workers failed to realise that lawfullness in social
as in physical science means an'if so' relation, a linkage between
hypothetical conditions and hypothetical effects. These laws do
not tell what conditions exist locally, at a given place at a given
time. In other words the laws don't do the job of diagnosis which
has to be done locally. Neither do laws prescribe the strategy for
change. (Lewin, 1946)
Of course Lewin is correct in all respects to object to the widespread
generalisation being applied in particular cases, but as I understand his
writing, he did not resolve this conceptual and value conflict in a way that
was appreciated by the minority groups and community councils that he
wanted to help through his insights and research.
These problems persist, indeed they are even more confounded today
in the plethora of interest networks and the rapidity of turnover of
information, from research to hype. Those that have the power to make
public definitions to realities have a far greater influence on social policy
than the sceptical, slow to judge researcher! Action research is not for the
impatient. When asked why he had not waited for the evaluation of the
participative Technical and Vocational Education Initiative the minister, Sir

11
 CLEMADELMAN
David Young, replied to the reporter that "one h a s to have faith in what one
does". Another 90 million pounds was then allocated. Both Lewin and
Young could be called pragmatists as they are both concerned with action
and consequences b u t Lewin seeks a n empirical basis for his arguments
whereas Young appeals to 'faith'.
Lewin is not a scientific positivist b u t a scientific pragmatist. His
methodology derives from C. S. Peirce, being a dialectical process seeking
best fit or concordance and an interpretative (of many social perspectives)
eplstemology melded to a quasi-experimental orientation. Lewin did not
work by hypothetical induction and objected to deduction in social science.
Nor was he a scientific realist obsessed with the promulgation of and
evidence for underlying laws. However, Lewin stressed the essential need to
formulate the hypothesis. Readers might wish to refer to recent articles in
Educational Researcher (Cherryholmes, 1992; House, 1992).
Argyris et al (1985) evaluate the contribution of Lewin and Dewey to
the founding of what they term "action science". In the quotes that follow it
is worth noting that the radical ideas of Lewin and Dewey remained largely
untried until the late 1960s.
Downloaded by [123.136.112.208] at 09:51 29 June 2015

Action science is an. outgrowth of the traditions of John Dewey

and Kurt Lewin. Dewey was eloquent in his criticism of the
traditional separation of knowledge and action, and he
articulated a theory of inquiry that was a model both for scientific
method and for social practice. He hoped that the extension of
experimental inquiry to social practice would lead to an
integration of science and practice. He based this hope on the
observation that 'science in becoming experimental has itself
become a mode of directed practical doing'.

This observation, that experimentation in science is but a special

case of human beings testing their conceptions in action, is at the
core of the pragmatist epistemology. For the most part, however,
the modern social sciences have appropriated the model of the
natural sciences in ways that have maintained the separation of
science and practice that Dewey deplored. Mainstream social
science is related to social practice in much the same way that the
natural sciences are related to engineering. This contrasts sharply
with Dewey's vision of using scientific methods in social practice.

One tradition that has pursued the integration of science and

practice is that exemplified by Lewin, a pioneer in group dynamics
and action research. Lewin is considered the founder of the
cognitive tradition within social psychology in America. Citing the
classic Lewinian studies of democratic and authoritarian group
climates, Festinger suggests that it is because Lewin showed how
complex social phenomena could be studied experimentally that

12
 THE ORIGINS OF ACTION RESEARCH

many regard him as the founder of modem experimental social

psychology. This is not to say, however, that each of the many
research programmes that can trace their core ideas to some
aspect ofLewin's work are also consistence with action science.
We consider Lewin himself to have been an action scientist
But since his time there has been a tendency to divorce his
contributions to science from those to practice. Research in social
psychology has relied on experimental methods for testing
hypothesized relationships among afew variables, and it has
become dtstantfrom practice. Practitioners in the applied
behavioural sciences, with some exceptions, have focused on
helping clients and have given little attention to testing scientific
generalizations.

The Lewinian tradition of action science, in contrast, is that of

scholar-practitioners in group dynamics and organizational
Downloaded by [123.136.112.208] at 09:51 29 June 2015

science who have sought to integrate science and practice.

Members of this tradition have emphasized the continuities
between the activities of science and the activities of learning in
the action context, the mutually reinforcing values of science,
democracy and education and the benefits of combining science
and social practice. (Argyris etal 1985)

Whatever the details of these important histories the 'American

Dream' of diversity and equal opportunity for aR was considered
to be in need of protection from the influence of the rising
totalitarian regimes of Europe. All means of research and
development were encouraged in the 1930s to the 1960s to foster
the 'democratic' rather than the 'autocratic' mentality in the home,
school and workplace. That is a long and in the main, yet to be
told, story of conflicting interests, finite resources and crumbling
theories.
Lewin and his workers classified their work into four types of action
research:
1. Diagnostic action research designed to produce a needed plan
of action. The change agents would intervene in an already
existing situation (for example, a race riot or anti-Semitic
vandalism), diagnose the problem, and recommend remedial
measures. Unless the proposed cures were feasible, effective, and
acceptable to the people involved, however, this design of action
was often wasted.

2. Participant action research in which it is assumed that the

residents of the affected community who were to help effect a cure

13
 CLEMADELMAN

must be involved in the research process from the beginning. They

would thereby realise mare keenly the need for the particular
steps finally decided upon; at the same time their 'ego investment'
would support the remedial program. This type of action research
- an example would be a community of self-survey - seemed to be
most effective for a limited range of problems. If was useful in
disclosing particular and local facts (not general principles) which
could provide examples for other communities.
3. Empirical action research was primarily a matter of record
keeping and accumulating experiences in day-to-day work,
ideally with a succession of similar groups, such as boys' clubs.
An inherent weakness of this procedure was that conclusions
were drawnfrom experience with a single group, or with several
groups differing in numerous ways, without test controls. Despite
this handicap empirical action research could lead to the gradual
development of generally valid principles as clinical medicine had
Downloaded by [123.136.112.208] at 09:51 29 June 2015

already demonstrated.

4. Experimental action research called for a controlled study of the

relative effectiveness of various techniques in nearly identical
social situations. Of all the varieties of action research, the
experimental had the greatest potential for the advancement of
scientific knowledge. Underfavourable circumstances it could
definitively test specific hypotheses. It was, however, the most
difficultform of action research to carry out successfully. (Marrow,
1969, p. 198)
Given Lewin's emphasis on participation we might expect this classification
to give emphasis to processes more than outcomes. As it is expressed the
classification is consistent with Lewin's search for axiomatic empirical
relationships. An argument for emphasis on process rather than outcomes
in participatory research may be found in Adelman & Fletcher (1982).
Lewin was unequivocal that action research could inform social
planning and action. Some recent UK authors have labelled the whole
process 'a cycle of action research1, whereas Lewin states that action
research may be only part of a process of social planning, reconnaissance
(evaluation of the action giving the planners a chance to leam the strengths
and weakness, so informing the next step and contributing to a basis for
overall modification of the planned change, or what has been latterly
identified with a cycle of action research), followed by review and iteration of
this overall cycle (Lewin, 1946).
By the time Lewin had established, in 1945, the Centre for Group
Dynamics Institute at the Massachusetts Institute of Technology his
colleagues included former students from Iowa, Festinger and Cartwright.
The chief methodological approach was to develop group experiments,
especially experiments of change to be carried on in the laboratory or in the

14
 THE ORIGINS OF ACTION RESEARCH
field. The Centre was to concern itself not only with the gathering of data
b u t with theorising that Lewin hoped would steadily keep ahead of the data
gathering. Lewin would wait until he perceived that the critical conditions
for a field experiment pertained before engaging in that work. He wanted his
'experiments' to be naturalistic yet interventive. The fundamental tenet was
studying things by changing them - in 'natural' situations.
Lewin could not rest on his successes but was in constant pursuit of
further funds for contracts and funding for research staff. He h a d
reluctantly acknowledged [pace Sanford, 1970) that action research was an
onerous and risky business, and that sponsorship for action research was
difficult to find. Eric Trist asked Lewin to act as consultant to a new
institute for the study of human relations in London; founded in part to
develop the discoveries about group conflict and cohesion, leadership and
influence for change, made during the Second World War by UK and North
American researchers in close co-operation. Lewin and Trist saw the
parallels in their ideas b u t to Lewin's regret he could not take up the offer.
Instead, one of his postdoctoral students, Eliot Jacques, went to help
establish the Tavistock Institute.
Downloaded by [123.136.112.208] at 09:51 29 June 2015

When Lewin died of heart failure in 1947, the Centre for Group
Dynamics, under the direction of Lewin's close associate Ronald Lippitt
(formerly a Director of Research with the American Boy Scouts), moved to
Michigan University at Ann Arbor. Lippitt's previous work in collaboration
with Lewin included the establishment, in 1945, of the National Training
Laboratories in Connecticut, which focussed particularly on sensitivity
training to combat radical and religious prejudice and racism. The Training
Laboratories drew upon the work of the Commission on Community
Inter-relations established through Lewin's persistence in 1944, with
sponsorship from the American Jewish Congress.
The pioneering action research of Lewin and his associates showed
that through discussion, decision, action, evaluation and revision in
participatory democratic research, work became meaningful and alienation
was reduced. Although power relations became more equitable in the
workplace this reconstructionist research made little difference to the
ownership of capital. Lewin and Dewey had similar ideas on participatory
democratic workplaces and schools b u t the institutionalisation of these
relationships h a s only been possible in parts of nations where wealth is
more evenly distributed, such as Norway fWirth, 1983). This has become
known as the 'quality of life' approach.

After L e w i n
I will now consider some subsequent UK action research that h a s
acknowledged its debt to Lewin and his associates' pioneering work. The
perceived merit of action research as a means to help solve social problems
by participative intervention has risen and fallen since the 1950s.
Currently, under the title of 'participative research' it is alive and well in the
UK, the USA and many other parts of the world, however, unlike the

15
 CLEMADELMAN

cautious public approach of Lewln, some present advocates are making

inflated claims for its impact on practice and policy and some are reifying
Individual development whilst neglecting the group and organisation. Lewln
had anticipated and criticised this tendency:
Recent research foldings have indicated that the ideologies and
stereotypes which govern inter-group relations should not be
viewed as individual character traits but that they are anchored
in cultural standards, that their stability and their change depend
largely on happenings in groups as groups. (Lewin, 1946)
During the 1960s UK and US social policy provided exceptionally large
budgets for Intervention programmes in education, health and housing.
These programmes were intended to raise the life chances, achievement
and expectations of the poor, otherwise called the 'disadvantaged' (Coates &
Silbum, 1970). At that time sociologists and psychometriclans were
confident enough in their knowledge of learning, social change and
organisation to eagerly suggest and participate in social engineering and
re-education programmes. The details of this history are to be found in
Downloaded by [123.136.112.208] at 09:51 29 June 2015

Silver & Silver (1991). Suffice to say here that the pioneering work of Tyler
in the assessment of learning and Lewin in the principles of co-operative
action research became urgently relevant and available through their
respective former students Benjamin Bloom and Martin Deutsch. In
education these interventions were termed 'compensatory' or 'enrichment'.
Although they were prone to justified criticisms then and subsequently (for
instance, Baratz & Baratz, 1970; Bernstein, 1970), at least those initiatives
went beyond the previously dominant determinist notions that the poor
could do little for themselves or were to be blamed for their faults and even
made to feel guilty for what was ascribed as their Inadequacies.
British educationalists, HMI, senior civil servants and politicians made
many study visits to the USA during the decade. Some were seeking
methodologies for systematic social development and new means of
evaluating the impact of public policy expenditure. Under the banner of
social action experiments the government funded Educational Priority Area
(EPA) and Community Development Projects (CDP) in England and Wales
(Halsey, 1972; Midwinter, 1972, 1975). Social reform was to be constructed
rationally using information coming out of the dialogue between social
science researchers and policy-makers. For the most part neither the EPA
or CDP projects proceeded by co-operative action research.
By the time the EPA project had begun in 1968, action research as a
means to cohesive social development had lost its coherence in the USA.
Instead of empowering ordinary people In their own communities, action
research had become incorporated as part of the armoury of managerial
development for "corporate excellence" (Blake & Mouton, 1968). Lewln's
ideas were so thoroughly digested and reformed as axioms, rather than
critically assimilated for further testing, that there is no reference to his
work in that and many other similar volumes of then and now, Lewln's
work on the understanding of intergroup conflict by means of the

16
 THE ORIGINS OF ACTION RESEARCH
community self-study was said by Rowan (1974) to be defunct while
Sanford (1970) claimed that action research was never accepted as bona
fide research in the USA.
It never really got off the ground, it never was widely influential in
psychology or social science. By the time the federal funding
agencies were set up after World War U, action research was
already condemned to a sort of orphan's role in social science -
for the separation of science and practice was now
institutionalised, and it has been basic to the federal
bureaucracies ever since. This truth was obscured for a time by
the fact that old timers in action research were still able to get
their projects funded: this after younger researchers had
discovered to their sorrow that action research proposals per se
received a cool reception from the funding agencies and were,
indeed, likely to win for their authors the reputation of being
'confused'.
Downloaded by [123.136.112.208] at 09:51 29 June 2015

The Humanities Curriculum Project (HCP; Stenhouse, 1975, 1980) engaged

participating teachers in the discussion of issues they identified from
classroom practice: the problems of implementing a humanities curriculum
which was itself based on pupil discussion, with the teacher acting as a
provider of resources and procedural chairperson. Although the project was
successful in many ways, teachers did not get the opportunity, as
recognised at the time, to make group decisions on change and Implement
these and evaluate the process and outcomes. However, the process of
introduction of the HCP strategies was evaluated under the rubric of
innovation by the HCP evaluation team led by Barry MacDonald (1978).
J o h n Elliott had been a member of the HCP curriculum team and in
his subsequent draft proposal to the Ford Foundation in 1971 highlighted
the need to follow through the problems of innovation and the realising of
pedagogies of enquiry and discovery in classrooms. The approach proposed
was that of action research. Elliott was quite clear about the need to engage
teachers in active participation and discussion but less clear about whether
decisions regarding further developments should be followed through by
individuals or by groups. It is worth a reminder that Lewin insisted that
action research was a group commitment. As well as focussing on
enquiry/discovery methods, Elliott at the outset suggested some of the
problems in curriculum areas and the methods by which these would be
researched.
I joined the Ford Teaching Project central team in March 1972. With
Rob Walker I had been working at the Centre for Science Education,
Chelsea College, on a Social Science Research Council project grant. We
found a few people who shared our developing ideas about school and
curriculum change. One of these was Barry MacDonald whom I heard on
a n Open University broadcast talking about the problems of evaluation. I
wrote to MacDonald and invited him to visit Walker and myself to see
samples of our work before our contracts concluded. In his reply he asked if

17
 CLEMADELMAN

he could bring his colleagues John Elliott. Subsequently Elliott sent me a

copy of his draft proposal to the Ford Foundation. In this it was contended
that although stimulating and challenging curriculum materials had been
devised, no such major change had been achieved in teachers' pedagogic
practices. The curriculum may be designed to foster enquiry through
independent reasoning but teachers were not articulating the means to
communicate these desired processes to students. There was an alarming
gap between the aspirations of education policy-makers, who decided on
expenditure for curriculum development, and the implementation of
programmes of curriculum change in classrooms. What Elliott, out of his
work with HCP, proposed was a project to enable teachers, through
collaborative action research on their own teaching, to make plain the
impediments to pedagogic change. The Humanities Curriculum Project had
begun this line of teacher based enquiry; the long-term observations of
Walker and I had raised similar questions about pedagogy; but neither had
the devotion to teachers' theorising and research as in what became known
as the Ford Teaching Project (Ford T).
I found then and subsequently (Adelman et al, 1983) that the most
Downloaded by [123.136.112.208] at 09:51 29 June 2015

difficult phase of action research was the preliminaries. To move from felt
'troubles' and 'anxieties' to a statement of an issue, teachers have to engage
in persistent reflexive thought about their own and others' practices. At
which point, often with help from the 'change agent', appropriate methods
for investigation of an issue can be suggested and constructed by
participants. It is at this point that the action research process begins to
come with the grasp of the participant researcher. However, prior to the
clarification there is a period of between a week to 3-4 months of awkward
talking around anecdotes and images trying to locate key actions and
acceptable terminology.
I found that participants' attempts to write down accounts of their
thoughts were of value in the process of reflective participant research. The
problem of Initial incoherence had nothing to do with the literacy or
intelligence of the Ford T or subsequently of other teachers. It seems to be
more to do with the gap between the ability of most people to perform
appropriate actions in an accomplished way and their ability to provide
descriptions of their own performances. This is a well-known problem In
psycholinguistics and ethnography, and it is also central to the work of
Donald Schon (1983).[2]
In the literature on educational action research, however, this vital
phase has been given far less attention than it deserves. The issue is often
presented as easily arrived at when the reality is quite the contrary. When
we asked groups of Ford T teachers to decide on which issue to explore in
their research there were various forms of consternation such as "we
thought you would tell use what we would research". Subsequent to initial
discussions about what was meant amongst the teachers about
enquiry/discovery teaching, 2 of the 40 teachers said they could no longer
be involved in the project because they no longer head the aspiration to
teach in that way.

18
 THE ORIGINS OF ACTION RESEARCH

Although teachers entered the project voluntarily and were supposed

to be aspiring to enquiry/discovery pedagogies, four teachers withdrew very
soon after the commencement of the project. Having received the first
documents from the central team they realised that the enquiry/discovery
pedagogies were not possible to implement in their schools. One withdrew
because the school was oriented to preparing children to pass formal
examinations, another thought that what he was doing was not
enquiry/discovery pedagogy but some form of guided instruction. A few
teachers voiced strong scepticism of the aims of the project in its initial
formulation. However, these sceptical teachers remained within the project
and were extremely valuable as critics and contributed a considerable
number of documents about their researchers. They became committed to
the project's aims, whilst reserving a detachment from drawing any firm
conclusions.
The teachers who had most difficulty in facing feedback from pupils,
from documents written by the project teachers or from discussions, were
those whose personal identity was inextricably bound up with particular
views of the professional role of a teacher. These teachers underwent
Downloaded by [123.136.112.208] at 09:51 29 June 2015

considerable stress, reported nightmares and insomnia and required extra

support from the central team. They stayed within the project, did write
documents, but developed much more slowly than the teachers who could
reflect on their own practices more readily.
The Ford T Project had sought democratic participation but found that
most of the 40 teachers were slow to participate actively; they waited to see
what developed with those teachers who were more ready to take the risk of
dissonance between their claims and their practices and the expression of
these in a public form. Although 'issues' from teachers were expected to
arise from their reflection of their 'troubles', the issues were often difficult to
express. At those junctures the participating teachers were vulnerable to
interventions by their colleagues, particularly those in a formal position of a
higher status, especially the headteacher. The central team could not be
sure that the issue was personal to the teacher and could not ensure, in
spite of numerous documents giving guidance meetings for discussion and
decision, that democratic, rather than autocratic, procedures would ensue.
Elliott and I could not monitor every meeting and the process of arrival at
the issues and their analysis. The teachers' case studies were intended to
provide such detail and the teachers' own research the systematic record.
We realised that Ford T had been successful in demonstrating that
teachers are able to research and theorise about their own practices. The
long reviews of Cook (1975 a, b) attracted even more attention to the Ford T
project. However, Elliott and I knew that a more pervasive and lasting
influence would come through explicit support of HMI. After their invited
visit to Norwich the response was that "It is not the policy of HMI to provide
funds for projects which they themselves have not initiated". It should be
noted that at about that time funds for dissemination were made available
to at least three DES/HMI projects in the areas of in-service education for
teachers (INSET) and school management.

19
 CLEMADELMAN

However, there was considerable interest in the project, particularly

from those in INSET and in local authority school advisory services. Some
chief education officers sent their representatives to conferences but HMI
did not ask Elliott to attend any of their national or regional conferences.
Further dissemination of the Project was mainly via the voluminous
writings of its participants. Productively, Elliott wrote overviews of these and
earlier Ford documents (see, for instance, Elliott, 1976) and these writings
attracted particular attention in North America (where three Ford teachers,
Elliott and myself gave seminars in New York, Chicago, Ohio and Toronto in
1976). The late Schools Council of England and Wales began to fund action
research in the two years before its demise in 1983. Steven Kemmis, a
former colleague at the Centre, for Applied Research in Education,
University of East Anglia (CARE), took educational action research to
Australia some four years after Rae Munro (1974) had begun similar work
in New Zealand.
This was the beginning of the Classroom Action Research Network
(CARN), which now has international membership (Adams, 1980; Somekh,
1990). Other regional initiatives in the UK include the Teacher Research
Downloaded by [123.136.112.208] at 09:51 29 June 2015

Network of Northern Ireland based at the University of Ulster and the Avon
Curriculum Review and Evaluation Programme. Like most programmes
these are small-scale and brief, with transitory funding. In an attempt to
further establish action research Jack Whitehead at the University of Bath,
Pamela Lomax at Kingston University and Richard Winter at Anglia
Polytechnic University supervise Master's courses which can include
dissertations based upon teachers' research into their own practice. These
studies are within the constraints of academic time and do not allow for the
risk associated with group participatory research. However, Colin Fletcher
at Cranfield Institute of Technology has since the 1970s developed
alternative ways of approaching the supervision of participatory research.
Sustained participatory research continues under the heading of
mutual support and observation (MSO) at Stantonbury Schools in Milton
Keynes (Fielding, 1989; Gates, 1989). In MSO three or more teachers
observe an issue in their mutual teaching and feed back this information to
each other. The observation, reporting and changes made are discussed
within the whole school amongst those who take part in MSO. These
comprise about 15% of the teachers in any one year. MSO has continued
since 1985.
In the UK a few places in England have sustained action research
through incorporating it into higher degree courses, as mentioned above,
and I have criticised this framing of the risky in the structure and
'progression' of academic courses (Adelman, 1989). CARN continues under
the guidance of Bridget Somekh at CARE. An annual international
conference and a bulletin are regular features. There are none of the
original group of Ford T teachers remaining in a membership of
approximately 400. CARN keeps the most complete list of the small projects
and dissertations in the UK.

20
 THE ORIGINS OF ACTION RESEARCH

The tendency to individual reflexivity using selective work of Schon as

the exemplar 14] rather than group research will not promote democratic
participation. Nor will the explicit yet convoluted distrust in teachers'
accounts as ideologically distorted misrepresentations of reality. As Elliott
(1991) argues, if claims are made to a distinction between 'practical' and
'emancipatory' action research, as is done by Carr & Kemmis (1986), they
should not deny the possibility of critical reflexive practices arising out of
the struggle by practitioners with their action research self-understandings.
The problem of participation is in the main who is to define the issue under
investigation, theorising about it and relationships in the process and in
whose name is the research publicised, if at all.
The lack of articulation with regional or national educational policy
formation has been commented on elsewhere (Carr, 1989). The means of
sharing vocabulary and meanings - the antidote to alienation - may be
through participatory research but the conditions for participation in that
research are hard won and harder to sustain. We in the UK may learn from
John Goodlad (Sirotnik & Goodlad, 1988), Herb Kohl (Kohl & Kohl, 1990),
Colin Fletcher (1988) and the PALM project in these respects.
Downloaded by [123.136.112.208] at 09:51 29 June 2015

I do believe action research, or rather participatory research, could be

a means to reconstruction (Simey, 1985) and productive work (Wirth,
1983). One of the urgent tasks is to bring together those who concentrated
on individual reflective practice such as Schon, those who try to carry on
Lewin's group discussion and decision and those that have worked with
large communities following the examples of Horton and Freire.
Participatory research may empower by raising the consciousness of
teachers about the social context in which they work, but participatory
research in its own right is still weak, lacking the sort of support that the
EPA and CDP projects briefly attracted. Currently, planning and decisions
over educational policy and practice are more and more being removed from
the local authorities to central government. Teachers are seen as operatives
in a system of line management; their work assessed and appraised, yet all
this without the local democratic politics of the North American School
Boards.
In this article I have argued that two deficient rhetorics have arisen
since Lewin: action research for greater effectiveness with, but more often
without, the link to democratic practice without sufficient or adequate
action research to demonstrate these claims. What has also been lost sight
of in the more recent emphasis on individual reflection has been the
essential inclusion of group and institutional relationships. Much of this
individual emphasis is attributed to the person-to-person consultant work
of Schon whilst his former co-author, Argyris, continues to investigate
organisation and group development in the Lewin tradition. The fruits of
reflexive thought, if they are claimed to have potential for improving practice
against stated criteria, have to be tested in joint and reciprocal social action
in the context of constraints and conflicts.
It remains to be seen whether participatory research can influence
social and educational policy in technocratic bureaucracies. There is every

21
 CLEMADELMAN

indication that in the UK the National Curriculum and assessment have in

no way been informed by participatory research.

Acknowledgements
My thanks to Professor Colin Fletcher, Dr Derek Purdy and Professor
Harold Silver for constructive criticism of the penultimate draft of this
paper. The author remains culpable.

Correspondence
Professor Clem Adelman, Faculty of Education and Community Studies,
University of Reading, Bulmershe Court, Earley, Reading RG6 1HY, United
Kingdom.

Notes
Downloaded by [123.136.112.208] at 09:51 29 June 2015

[1]Alfred J. Marrow, his biographer, was the secretary to the American committee.
[2] Whose individualistic rather than group approach to development comprises but
one of the differences between his work and participatory research. This may be a
reason why Carr & Kemmis (1986) do not mention his work, albeit this absence is
reciprocal.

Bibliography
Adams, E. (1980) Ford Teaching Project, in L. Stenhouse (Ed.) Curriculum Research and
Development in Action. London: Helnemann.
Adelman, C. (1989) The practical ethic takes priority over methodology, in W. Carr (Ed.)
Quality in Teaching: arguments for a reflective profession. Brighton: Falmer Press.
Adelman, C. & Fletcher, C. (1982) Collaboration as a research process, Quarterly
Journal of Community Education, 1, pp. 15-24.
Adelman, C , Boxall, W., Parson, I., Thebault, Y., Treacher, T. & Richardson, R. (1983)
A Fair Hearingfor Alb relationships between teaching and racial equality. University
of Reading, Bulmershe Research Publication No. 2.
Argyris, C. & Schon, D.A. (1978) Organizational Learning. Reading, MA:
Addison-Wesley.
Argyris, C., Putnam, R. & McLain Smith, D. (1985) Action Science. San Francisco:
Jossey-Bass.
Baratz, S.S. & Baratz, J.C. (1970) Early childhood intervention: the social science base
of institutional racism, Harvard Educational Review, 40, pp. 29-50.
Bernstein, B. (1970) Education cannot compensate for society, in Language in
Education. London: Routledge & Kegan Paul (in association with the Open
University Press).
Blake, R.R. & Mouton, J.S. (1968) Corporate Excellence throughGrid.Organization
Development a systems approach Texas: Gulf Publishing Co.
Carr, W. (1989) Understanding quality in teaching, in W. Carr (Ed.) Quality in
Teaching: arguments for a reflective profession. Brighton: Falmer Press.

22
 THE ORIGINS OF ACTION RESEARCH

Carr, W. & Kemmls, S. (1986) Becoming Critical: education, knowledge and action
research. Geelong: Deakin University Press.
Cartwright, D. & Zander, A. (1953) Croup Dynamics. London: Tavistock.
Chenyholmes, C.H. (1992) Notes on pragmatism and scientific realism, Educational
Researcher, 21(6), pp. 13-17.
Coates, K. & Silburn, R. (1970) Poverty - the forgotten Englishman. Harmondsworth:
Penguin.
Cook, M. (1975a) Where the action research is - a look at the innovatory work arising
out of the Ford Teaching Project, The Times Education Supplement, 11 July.
Cook, M. (1975b) Bridging the gap between theory and practice - a review of Ford
Teaching Project publications, The Times Educational Supplement, 18 July.
Corey, S.M. (1953) Action Research to Improve School Practices. New York: Bureau of
Publications, Teachers' College, Columbia University Press.
Elliott, J. (1976) Developing hypotheses about classrooms from teachers' practical
constructs, Interchange, 7(2), pp. 2-22.
Elliott, J. (1991) Action Researchfor Educational Change. Milton Keynes: Open
University Press.
Elliott, J. & Adelman, C. (1975) Teacher education for curriculum reform: an interim
Downloaded by [123.136.112.208] at 09:51 29 June 2015

report on the work of the Ford Teaching Project, British Journal of Teacher
Education, 1, pp. 105-114.
Fielding, M. (1989) The fraternal foundations of democracy: towards emancipatory
practice in school-based INSET, in C. Harber & R. Meighan (Eds) The Democratic
School Ticknall: Education Now Publishing Co-operative.
Fletcher, C. (1988) Issues for participatory research in Europe, Community
DevelopmentJournal,23, pp. 44-46.
Gates, P. (1989) Developing consciousness and pedagogical knowledge through mutual
observation, in P. Woods (Ed.) Working for Teacher Development. Derham: Peter
Francis.
Halsey, A. (1972) Educational Priority, Vol. 1. London: HMSO.
House, E. (1992) Response to 'Notes on pragmatism and scientific realism', Educational
Researcher, 21(6), pp. 18-19.
Kohl, H. & Kohl, S. (1990) The Long Haul - an autobiography of Myles Horton. New York:
Doubleday.
Landsberger, H.A. (1958) Hawthorne Revisited: management and the worker, its critics
and developments in human relations in industry. Ithaca: Cornell University Press.
Lewin, K. (1920) Die Sozialisierung des Taylorsystems, Praktischer Sozialismus, No. 4.
Lewin, K. (1946) Action research and minority problems, in G.W. Lewin (Ed.) Resolving
Social Conflicts. New York: Harper & Row (1948).
Lewin, K., Lippett, R. & White, R.K. (1939) Patterns of aggressive behaviour in
experimentally created social climates, Journal of Social Psychology, 10,
pp. 271-301.
MacDonald, B. (1978) The Experience of Innovation. Norwich: CARE, University of East
Anglia.
Madaus, G.F. & Stufflebeam, D. (1989) Educational Evaluation: classic works of Ralph
W. Tyler. Boston: Kluwer.
Marrow, A.J. (1969) The Practical Theorist the life and work of Kurt Lewin. New York:
Basic Books.

23
 CLEMADELMAN
Midwinter, E.C. (1972) Priority Education. Harmondsworth: Penguin.
Midwinter, E.C. (1975) Education, and Community. London: George Allen & Unwin.
Munro, R.G. (1974) Self-monitoring teachers, The Times Educational Supplement, June.
Rowan, J. (1974) Research as an intervention, in N. Armistead (Ed.) Reconstructing
Social Psychology. London: Penguin.
Sanford, N. (1970) Whatever happened to action research? Journal of Social Issues,
26(3).
Schon, D.A. (1983) The Reflective Practitioner. New York: Basic Books.
Silver, H. & Silver, P. (1991) An Educational War on Poverty: American and British
poUcy-making, 1960-1980. Cambridge: Cambridge University Press.
Simey, M. (1985) Government by Consent the principle and practice of accountability in
local government London: Bedford Square Press.
Simons, H. & Elliott, J. (1989) Rethinking Appraisal and Assessment Milton Keynes:
Open University Press.
Sirotnik, K.A. & Goodlad, J.I. (1988) School-University Partnerships in Action. New
York: Teachers' College Press.
Smith, G. (1987) Whatever happened to educational priority areas?, Oxford Review of
Downloaded by [123.136.112.208] at 09:51 29 June 2015

Education, 13, pp. 23-39.

Somekh, B. (1991) Pupil Autonomy in Learning with Microcomputers: rhetoric or reality?
An Action Research Study. Norwich: PALM Publications.
Stenhouse, L. (Ed.) (1975) An Introduction to Curriculum Research and Development
London: Heinemann.
Stenhouse, L. (Ed.) (1980) Curriculum, Research and Development in Action. London:
Heinemann Educational.
Wirth, A.G. (1983) Productive Work in Industry and Schools. New York: Universities
Press of America.
Wright Mills, C. (1959) The Sociological Imagination. New York: Oxford University Press.

24